Accessibility statementSkip to main content
We need your help: All banner donations made today will support The Daily's new staff financial aid program.
Learn more and donate.


Research Roundup: Shelter in place impacts, forest fire predictions, poverty level estimates


Each week, The Daily’s Science & Tech section produces a roundup of the most exciting and influential research happening on campus or otherwise related to Stanford. Here’s our digest for the week of May 17-23.

Shelter in place impact on carbon dioxide emissions

Daily carbon dioxide emissions have dropped by 17% globally as a result of shelter in place policies, found a study published on May 19 in “Nature Climate Change.”

“The drop in global emissions we estimate this year will surprise some people in being ‘only’ 4-7% because shelter-in-place rules are temporary and staggered across different countries,” earth system science professor Rob Jackson told Stanford News. “But it will still be the biggest emissions drop since World War II, though for undesirable and unsustainable reasons. More surprisingly, U.S. emissions declined one third for part of April, a shocking drop driven by reduced mobility, manufacturing and electricity demand.”

The findings suggest that the changes in emissions may be temporary because they are not caused by structural changes to the economic, transportation or energy systems. According to the article, the length of shelter-in-place policies may have wide range impacts on the 2020 annual carbon dioxide emissions.

If shelter-in-place policies are lifted by mid-June, low estimates indicate a 4% decrease in annual emissions. If confinement remains the global standard until the end of 2020, researchers expect a high estimate of 7% decreased annual emissions.

AI predicts forest fires amid fire season

Deep-learning models and artificial intelligence (AI) can help scientists predict forest fires as many western states head into fire season, a study published on May 8 in “Remote Sensing of Environment” reports.

“One of our big breakthroughs was to look at a newer set of satellites that are using much longer wavelengths, which allows the observations to be sensitive to water much deeper into the forest canopy and be directly representative of the fuel moisture content,” earth system science assistant professor Alexandra Konings told Stanford News.

The new model uses a recurrent neural network: AI that can recognize patterns within large amounts of data. Researchers trained their deep-learning model using data from the National Fuel Moisture Database. The model can estimate live fuel moisture levels, a well known quality that influences wildfire risk, and create an interactive map that fire agencies use to fight fires.

“Creating these maps was the first step in understanding how this new fuel moisture data might affect fire risk and predictions,” Konings told Stanford News. “Now we’re trying to really pin down the best ways to use it for improved fire prediction.”

AI estimates poverty levels and development changes

Satellite imagery and artificial intelligence can help researchers estimate poverty levels and changes in infrastructure development over time in African villages, a study published on May 22 in “Nature Communications” reports.

Government agencies and organizations can use the information to deliver services and necessities to aid people.

“Amazingly, there hasn’t really been any good way to understand how poverty is changing at a local level in Africa,” earth system science professor David Lobell told Stanford News. “Censuses aren’t frequent enough, and door-to-door surveys rarely return to the same people. If satellites can help us reconstruct a history of poverty, it could open up a lot of room to better understand and alleviate poverty on the continent.”

The satellite images contain photos from both daytime and nighttime. During the day, images of human infrastructure correlate with development; nighttime images that contain lights correspond to development as well.

The AI deep learning model analyzes satellite imagery to form an asset wealth index, an economic measurement used to evaluate household wealth.

“Our big motivation is to better develop tools and technologies that allow us to make progress on really important economic issues. And progress is constrained by a lack of ability to measure outcomes,” earth system science assistant professor Marshall Burke told Stanford News. “Here’s a tool that we think can help.”

While you're here...

We're a student-run organization committed to providing hands-on experience in journalism, digital media and business for the next generation of reporters. Your support makes a difference in helping give staff members from all backgrounds the opportunity to develop important professional skills and conduct meaningful reporting. All contributions are tax-deductible.


Get Our EmailsGet Our Emails